Improving business processes : A case study of AS Spilka Industri

(1)

Master’s degree thesis

LOG950 Logistics

Improving business processes:

A case study of AS Spilka Industri

Maria Aklestad Garnes Tonje Vikhagen

Number of pages including this page: 111

Molde, May 2011

(2)

Publication agreement

Title: Improving business processes: A case study of AS Spilka Industri Author(s): Maria Aklestad Garnes and Tonje Vikhagen

Subject code: LOG950 ECTS credits: 30 Year: 2011

Supervisor: Bjørn Guvåg

Agreement on electronic publication of master thesis

Author(s) have copyright to the thesis, including the exclusive right to publish the document (The Copyright Act §2).

All theses fulfilling the requirements will be registered and published in Brage HiM, with the approval of the author(s).

Theses with a confidentiality agreement will not be published.

I/we hereby give Molde University College the right to, free of

charge, make the thesis available for electronic publication: yes no

Is there an agreement of confidentiality? yes no

(A supplementary confidentiality agreement must be filled in)

- If yes: Can the thesis be online published when the

period of confidentiality is expired? yes no

Date: 24 May 2011

(3)

I

(4)

II

Acknowledgements

The authors are highly thankful to all the people who have helped during the work with this thesis.

The project has been carried out in co-operation with AS Spilka Industri. Spilka has opened the company to us, allowing us to visit anytime and providing us with an office, and for this we are grateful. We would especially like to thank Managing Director Leif-Erling Løvoll and Plant Manager Geir Lervåg for all the interesting ideas and discussions, and for providing us with the data and information we have needed. We are also thankful to the employees at the office and warehouse at Spilka for being open-minded about the thesis, for guiding us in the process, for answering questions, and for sharing ideas and comments.

We would like to show our gratitude to supervisor Bjørn Guvåg whose guidance and feedback have been of high value. Trough the process he has contributed with inspiring ideas and encouraging support.

Molde, May 2011

Maria Aklestad Garnes Tonje Vikhagen

(5)

III

Abstract

This paper is a co-operation project with AS Spilka Industri, a producer of hinges and fittings for windows. The focus of the thesis is on the business process of handling customer orders within this company and how it can be improved. The purpose of the project was to identify problems in the order process and investigate the root causes of these. Last, the intention was also to define and measure the delivery performance.

Lean and Six Sigma are introduced as the main theoretical concepts in this thesis, giving a philosophical perspective in the form of principles and goals as well as a practical perspective in the form of tools and techniques. They are both considered state-of-the-art methodologies and provide valuable tools for process improvement that are used in the analysis.

This thesis has applied a method for measuring lead times in business processes that required employee involvement. The collection of lead time data has been achieved by using lists where the people involved in the process have recorded lead times. The units of analysis have been limited to orders for standard products.

The analysis is divided into three parts. The first part concentrates on defining, describing and analysing the order process and the associated problems using Lean tools. The next part centres on value stream mapping and a method for how the current situation can be described and measured. Finally a metric and a tool for measuring delivery performance have been developed.

The paper focuses on two main problems in the order process: errors in deliveries and part deliveries. By using Lean tools the causes of these errors are explored further. The results from measuring of lead times have been divided into value adding and non-value adding time, and it shows that non-value adding waiting time stands for a large share of total time.

The analysis of the current order process leads to recommendations to Spilka on how they can reduce waste and improve flow in the order process.

(6)

IV

List of figures

Figure 1 The three production systems (The Automotive Consulting Group, 2011) ... 10

Figure 2 Illustration of value added and non-value added activities (MacMahon, 2009) ... 14

Figure 3 the 5 S’s (adapted from Liker, 2004) ... 18

Figure 4 A System View (adapted from Dennis, 2002) ... 19

Figure 5 Cause-and-effect diagram (Management Systems Inc, 2006) ... 20

Figure 6 Value Stream Mapping Icons (adapted from an overview in Keyte and Locher (2004)) ... 23

Figure 7 DMAIC model (Lean Sigma institute, 2010) ... 28

Figure 8 The relationship between the five Lean principles and DMAIC (Salah, Rahim and Carretero, 2010) ... 30

Figure 9 Pareto chart of delivery errors ... 49

Figure 10 Cause-and-effect diagram for root causes of errors ... 53

Figure 11 The development of part deliveries over the last 4 years ... 56

Figure 12 Annual sales at Spilka over the last 4 years, based on internal data ... 57

Figure 13 Product lines delivered for each month (2008-2010), based on internal data ... 58

Figure 14 Product lines delivered for each week (2008-week 18, 2011), based on internal data ... 61

Figure 15 Working hours ... 62

Figure 16 Value Stream Map for domestic orders ... 66

Figure 17 Value Stream Map for international orders ... 67

Figure 18 Lead times ... 68

Figure 19 Lead time divided into activities for domestic orders ... 69

Figure 20 Lead time divided into activities for international orders ... 70

Figure 21 Lead time divided into value adding and non-value adding activities ... 71

Figure 22 Value adding time ... 72

List of pictures

Picture 1 Spilka Classic hinge (Spilka, 2011a) ... 2

Picture 2 Spilvent (Spilka, 2011b) ... 2

Picture 3 Spilka’s marketing and production facility in Ålesund (Spilka, 2011c) ... 3

List of tables

Table 1 Basic Research Design (Ellram, 1996) ... 34

Table 2 Primary Data (Hox and Boeije, 2005) ... 37

Table 3 Overview of data collected in the project ... 39

Table 4 The products with the most part deliveries in 2010. ... 56

(10)

1

1. Introduction

This thesis focuses on improving business processes, and especially the order process. It is carried out as a case study of AS Spilka Industri. For simplicity, AS Spilka Industri will often be referred to as just Spilka in the remainder of the paper. The order process can be defined shortly as the sequence of activities that are associated with the filling of customer orders.

The process will in the thesis be elaborated and analysed on the basis of theoretical concepts such as Lean and Six Sigma.

The paper is divided into six main chapters. The first chapter presents the company, AS Spilka Industri, and the research problem for the thesis. Chapter 2 focuses on the theoretical framework, giving an overview of the concepts of Lean and Six Sigma, before the research methodology and data collection methods are explained in chapter 3. The analysis is carried out in chapter 4, centring on the order process, value stream mapping and delivery

performance. Chapter 5 presents a conclusion and recommendations to the company, while limitations of the study and further research are presented in chapter 6.

1.1. AS Spilka Industri

The thesis will take a closer look on AS Spilka Industri, a company located in Ålesund at the north-western coast of Norway. The company is now the world leading producer of hinges and fittings for the top hung fully reversible windows by H-Window. The next pages will give an introduction to the company. All information is retrieved from Spilka’s web site or given to the authors by Spilka.

Spilka has a long history of production. They originally produced baby carriages and buggies under the name Spjelkavik Barnevognfabrikk AS, but after 15 years of operation, in 1948, they shifted to production of hinges and fittings for the local furniture industry. Shortly after this, the company was given its current name, AS Spilka. In 1958, the local inventor Harald Kvasnes developed the first fully reversible hinge for use on windows. This meant that windows could be opened and reversed, making it possible to clean the outer glass pane from inside. The window was named “Husmorvinduet” (Housewife’s Window), later known as just “H-Vinduet” in Norwegian and “H-Window” in English. Harald Kvasnes and Spilka

(11)

2 formed a partnership, and Spilka started to produce and sell the hinges.

Picture 1 Spilka Classic hinge (Spilka, 2011a)

During the years the hinge has been redesigned and improved, and more products have been developed. Since the 1980s all of Spilka’s production has been window-related products. They can offer hinges and other components for top hung fully reversible windows (Classic), side hung fully reversible windows (Swing), sliding patio doors (Tango), and aluminium-clad top hung fully reversible windows (Opus). All hinges come in different sizes depending on the size of the window. Classic is now the name of the original hinge for the H-Window, which still is the main product for Spilka. The Classic hinge is shown in picture 1. Spilka owns two registered trademarks. In addition to H-Window they have developed Spilvent, which is a ventilator that can be used with most window types. Spilvent is shown in picture 2.

Picture 2 Spilvent (Spilka, 2011b)

(12)

3 Spilka’s products are sold to customers in Norway as well as exported to other countries around the world. But the main market is in the United Kingdom and in Scandinavia. As of 2010 the company has an annual turnover of about NOK 120 million. Based on 2010 sales figures it can be seen that Spilka’s sales are fairly equally distributed between domestic and international customers. 52 % of Spilka’s customers are Norwegian, and the Norwegian customers stand for 52 % of the sales. The major customers are window producers, thus Spilka is operating in the business-to-business market. Good and long-term relationships with customers are of high importance for Spilka, and so they value timely and correct deliveries as well as high quality on the products.

The marketing and production department is located in the same facilities in Ålesund, and this is where the authors have visited to do research for this thesis. The production facilities in Ålesund consist of two production halls, each of them having a total production area of 1900 m². Additionally there is a branch office of the marketing department in the UK covering the Polish and Baltic markets.

Picture 3 Spilka’s marketing and production facility in Ålesund (Spilka, 2011c)

Spilka also has a Research and Development (R&D) department with 5 employees, which is placed at a different location in Ålesund. This department is responsible for development and design, testing of prototypes, technical support and production control. More than 5%

of the company’s annual turnover is used to improve their existing products and also to develop new innovative products (Spilka, 2011d).

(13)

4

1.2. Research Problem

In this section of the thesis the research problem will be outlined and presented. First it will describe the background for the project, followed by an explanation of the research

problem. Finally we have narrowed down the research problem into the formulation of research questions.

Spilka contacted Molde University College in September 2009. The company wanted students with a higher level of education to study activities in their company. They initially wanted the students to focus on the flow of goods, receiving inspections, packing and forwarding. It was decided by the company that one would not look into the production areas of the company in this project. Spilka has previously focused much on their production processes and made improvements in this area, and it was in their opinion that other areas of the company had better opportunities for improvement. In May 2010 an agreement was made between the authors and Spilka about a master thesis project for the spring semester of 2011.

In 2010 the company started a project on customer satisfaction. The objective of this project was to identify areas that the customers are not satisfied with. Spilka experiences that some of their customer orders cannot be delivered completely because not all products are available when the order must be sent, and that they therefore have to use part deliveries.

Feedback from customers showed that one of the largest weaknesses is that the customer receives information too late when orders cannot be delivered as promised. Customers have not always been informed when products are missing from the shipment until delivery.

Further Spilka thinks that the number of errors in deliveries is no longer satisfactory. We decided to use these observations as a basis for the further development of the research problem.

The problems with information and errors in deliveries that Spilka has recognized are parts of business processes in the company. Monk and Wagner (2009, p. 3) define a business process as “a collection of activities that takes one or more kinds of input and creates an output that is of value to the customer”. This customer might be internal in terms of other business activities depending on the process or external in the form of a traditional

(14)

5 customer that buys the product. Business processes cross functional areas in an

organisation, such as sales and marketing, supply chain management, HR and accounting.

Each functional area includes a number of business functions, which are activities that are performed under the functional areas. Supply chain management, for example, consists of purchasing, production, transportation and receiving goods (Monk and Wagner, 2009). All businesses consist of a number of business processes (Willoch, 2005).

The business process in Spilka that has been most affected by the problems mentioned above is the process of handling customer orders. When products are missing from the delivery there has been an error somewhere in the activities from the order was registered to the final ordered being shipped, and the customer is not satisfied. Therefore, the

business process we will concentrate our research on is the order process. Willoch (2005) describes a process that he calls “filling of orders”, which is similar to the one referred to as order process in this thesis. This process starts when the customer has a need for a product, and ends when the customer receives the product. Such a process exists in all businesses that produce or trade goods, and it typically crosses business functions such as sales and marketing, forecasting, production planning, inventory management, distribution etc. Other definitions (Samaranayake, 2009) also include pre-sales activities with the goal of giving price information to the customer. Such activities include sales calls and visits, and these are the first activities in the customer order process. Sales orders are created on basis of the price information given in the pre-sales activities. This is followed by other order process activities such as inventory sourcing (checking if the products needed are available), documents release, picking and packing, distribution planning and invoice creation and customer payment.

In this thesis we have used these explanations of the order process and its activities and defined it so that it suits the way the order process is performed at Spilka. First we will present a short definition of the order process here, and then it is going to be described in more detail under chapter 3 about the research methodology for the thesis. The order process can be defined as the sequence of activities that are associated with the filling of customer orders. It uses customer orders as input, and the output is the physical goods

(15)

6 being shipped. The functional areas in Spilka that are involved in this process are the order department and the warehouse, and those are the functions that we intend to work with.

When solving a research problem it is important to define interesting research questions that should be answered. According to Yin (1994) the process of defining the research questions is probably the most important step to be taken in a research study. The process of determining the research questions requires much preparation, so that the questions are precise and significant for the topic.

In this study we want to identify problems that Spilka has in the order process, and find the root causes of these problems. Further we want to measure and analyse the problems to describe how they are doing today. By identifying the problems Spilka has in the order process, suggestions on how to improve the problems and the business process will be made. The aim is to find out how Spilka can improve their business processes, using the order process as a main focus. During the work with the project it was discovered that though Spilka has an increasing focus on improving customer satisfaction, they have not yet clearly defined delivery performance and how they wish to perform in this area. It was therefore decided to include this in the research problem in order to identify Spilka’s meaning of the concept and to develop a metric that can be used later to see the development. These elements can be summed up in three research questions.

Research questions

- What problems can be identified as the main problems in the order process at AS Spilka Industri, and what are the root causes of these problems?

- How can the delivery performance be defined and measured?

- How can the business process at AS Spilka Industri be improved to remove waste and increase delivery performance?

(16)

7

2. Theoretical framework

In this chapter an outline of the relevant theories related to the research problem will be given. The first part presents the theoretical concept of Lean production. Here the main principles and the origin of Lean will be presented, followed by a description of the differences between Lean production and other production systems. Next there is an explanation of the elements in the Lean philosophy. Within Lean there are several tools and techniques, and this chapter will introduce some of them, including 5S, 4M, 5 whys, cause- and-effect diagrams and value stream mapping. The Six Sigma concept will also be

described, followed by one of its main problem solving methodology, DMAIC. Finally the chapter will present a brief look at the relationship between Six Sigma and Lean.

Lean and Six Sigma are relevant for the project, since both are methodologies that provide tools for how to improve processes. They are used by many companies in various industries and are referred to as state-of-the-art methodologies for process improvement (Salah, Rahim and Carretero, 2010). Lean focuses on the elimination of waste and Six Sigma focuses on improving quality and efficiency.

2.1. Lean production

The term Lean production was introduced by John Krafcik, one of the researchers in the International Motor Vehicle Program (IMVP) (Womack, Jones and Roos, 2007). This was a research program initiated at the Massachusetts Institute of Technology (MIT) in 1985, aiming at making a survey of the car production industry worldwide. The study included companies and plants in 14 countries over a time period of 5 years. The term Lean became popular after it was used in the book The Machine That Changed the World by Womack, Jones and Roos, in which the findings of the research program were presented. This book thoroughly describes a Lean system, but there is no explicit definition of Lean production (Shah and Ward, 2007). Plenert (2007) states that the concept of Lean has had many names, including Toyota Production System, Just in Time, Pull Manufacturing and Total Quality Management. Further he claims that Lean today is a collection of tools and methodologies, and when working with Lean improvement it is essential to establish a mix of appropriate tools in order to achieve the organisation’s objectives. One definition of Lean may be this

(17)

8 from the Lean consulting company MainStream Management, quoted in Plenert (2007, p.

146):

“Lean is a systematic approach that focuses the entire enterprise on continuously improving quality, cost, delivery, and safety by seeking to eliminate waste, create flow, and increase the velocity of the system’s ability to meet customer demand.”

Lean production can be described from two perspectives. There is a philosophical

perspective that involves a way of thinking, in terms of guiding principles and overarching goals, and then there is a practical perspective that includes management practices, tools and techniques (Shah and Ward, 2007). Shah and Ward (2007, p. 791) propose the following definition in order to encompass many of the different elements of Lean:

“Lean production is an integrated socio-technical system whose main objective is to eliminate waste by concurrently reducing or minimizing supplier, customer, and internal variability”

This variability that companies have to manage may be variability in supply, processing time, or demand.

2.1.1. Lean principles

There are five fundamental principles of Lean. These are described in Bicheno (2004).

 Specify value from the point of view of the customer. Products and processes should be designed based on the needs of the customers and not on what the company finds convenient.

 Identify the value stream. A value stream can be defined as a set of operations from raw material to the final customer. This should be mapped (see 2.3.5. about value stream mapping).

 Flow. The aim is to have a good flow in the process, so that there are no queues and delays. Especially a value-adding step should not be delayed by a non-value adding

(18)

9 step; in that case one should try to organise the process differently. Lean encourages the idea of “one piece flow” in operations. This means sending single parts or

products or very small lots of them from one operation to another within a cell consisting of people, machines and workstations grouped closely together (in a processing sequence) (Liker, 2004).

 Pull. This principle involves producing only as a response to downstream demand, either from a final customer or from an internal customer. Lean aims at moving the point where push changes into pull further upstream in the process. If operations work at the demand rate of the final customer, overproduction, one of the seven wastes, can be avoided (see overview of the seven wastes in section 2.2.2.).

 Perfection. This can be achieved if the principles above are fulfilled. Perfection means zero waste.

These are goals within the Lean philosophy that may be impossible to fully achieve.

However, they are part of a vision which one can work towards and with that improve by reducing waste.

2.1.2. Difference between mass production and Lean production

Lean was created as a term because it uses less than mass production: fewer people are needed, less manufacturing space, fewer machines and tools, less inventory, fewer defects and less time to develop new products. Lean has the goal of perfection, even if it probably will not be reached. Perfection implies falling costs, zero defects, zero inventories and high product variety (Womack, Jones and Roos, 2007).

The literature points out three main types of production systems: craft production, mass production, and Lean production (Womack, Jones and Roos, 2007 and Krafcik, 1988). Craft production appears first historically and is characterized by skilled workers that are involved in the production process of the whole product, not just a part of it. They use general- purpose tools and have very low production volumes (Womack, Jones and Roos, 2007). Two car manufacturers may illustrate the difference between mass production and Lean

production. Often Ford is used as a typical example of a mass producer while Toyota is the typical example of a Lean producer. However in recent years the difference between the

(19)

10 production systems of these two companies has decreased (Krafcik, 1988). Figure 1 below shows the three production systems in a matrix. Craft production achieved high

differentiation, but the move to a mass production system enabled the industry to produce at a lower cost. Finally, Lean production seeks to achieve both high differentiation and low cost (The Automotive Consulting Group, 2011).

Figure 1 The three production systems (The Automotive Consulting Group, 2011)

Krafcik (1988) describes the difference between production systems using characteristics such as span of worker control, inventory levels, and size of repair areas.

Span of worker control:

In mass production, the workers have only one or two narrowly defined tasks to perform and they have little span of control over the finished product. The tasks are simple and of short duration so they are repeated several times a day. This standardization of tasks reduces the time needed for employee training. Toyota also used standardized work, but they made the employees responsible for standardizing the work and for continuously improving performance. They used the idea from craft production that workers should be

(20)

11 skilled, combined it with the standardized work and assembly line of Ford’s system and added team work as an important element (Krafcik 1988).

Part inventory levels:

The mass producers’ way of achieving low production costs per unit is through economies of scale. This implies large batch sizes, which lead to high inventory levels, both of raw

materials and components as well as work in process and finished parts (Womack, Jones and Roos, 2007). Lean production systems may be called high risk-high return systems. The inventory buffer is low, so if something goes wrong the production will come to a halt because there is no extra inventory or utility workers. The risk can however be minimized with well-trained workforce, responsive suppliers and good product designs. Systems that rely on high buffers have lower risk, but also lower potential for high performance (Krafcik, 1988).

Repair areas:

Mass production accepts that there might be a need for rework after the product is finished.

Mistakes can be corrected before shipping the product to the customer, since it would be costly to stop the whole production line just because of a small error. Because mass production relies on large batch sizes, a defect will most likely not be discovered before many parts with the same defects are produced, thus the need for rework. Lean aims at discovering errors and defect in the process and immediately notifying those responsible so that the cause can be found and improved at once (Womack, Jones and Roos, 2007).

2.1.3. Toyota Production System

Lean is based on the Toyota Production System (TPS), a system developed by the leader of Toyota in the years after WWII, Eiji Toyoda, and his plant manager, Taiichi Ohno. After a tour of Ford’s car manufacturing plant in the US, Toyoda realized that the Japanese market was too small to produce as many cars as Ford did, and they also did not have enough money to invest in as many machines. Ford used a large number of machines that each specialized in manufacturing one part in large batches, reducing costs per unit. Toyota

(21)

12 therefore had to achieve higher productivity through a more flexible process (Liker, 2004).

This was the beginning of Toyota Production System.

At the time of Eiji Toyoda’s visit, Ford’s plant Rouge was the world’s largest and most efficient manufacturing facility. However, changeovers from producing one car model to producing another could take a long time. In 1927 Ford kept one of its plants closed for months when it switched from Model T to Model A (Womack, Jones and Roos, 2007). One of the methods that Ohno and Toyota decided to improve was the time it took to change dies on a machine. (Dies are used in a machine to press sheets of steel into a shape that is needed in the car.) Usually changing them took a full day, so large car producers did not do this very often. For Toyota that did not have that high production volume or many

machines, the dies had to be changed more often. Ohno was eventually able to develop a technique to easily change the dies using only three minutes. In the process he also

discovered that the cost per part of producing in small batches actually was lower than for producing in large batches. The main reasons for this were lower inventory costs and that defects were discovered almost immediately (Womack, Jones and Roos, 2007).

Another key aspect of the Toyota Production System is the focus on the employees. Toyota experienced some difficulties due to a depression in the late 1940s, and they had to

discharge one fourth of their workforce. In this process however, Toyota made a deal with the remaining employees. First the president of Toyota, Kiichiro Toyoda, resigned to show responsibility for the problems. Second, the workers were guaranteed lifetime employment and salaries that increased with seniority. In return employees would work to help improve the company (Womack, Jones and Roos, 2007).

Some of the ideas behind TPS came from the US. One of those was the pull principle, inspired by the replenishment of items on shelves in American supermarkets. The pull principle has been explained as one of the Lean principles in section 2.1.1. TPS, or Lean, started with the Toyota company, but was soon spread to also include Toyota’s many suppliers and dealers (Liker, 2004).

(22)

13

2.2. Elements of the Lean philosophy

2.2.1. Value and waste

There are different views on how value should be measured. Many agree however on the point that value has to be defined from the customer’s point of view (Dennis, 2002). Plenert (2007, p. 285) defines value as “A capability provided to a customer at the right time at an appropriate price, as defined in each case by the customer”. Bicheno (2004) emphasizes that the future value also may be considered. This is what the future customers are willing to pay for, that the current customers do not value as much. However, this aspect is more relevant for research and development than for the current production.

The first question in TPS and Lean is: What is the value of the process to the customer? This applies to both internal and external customers. Internal customers are activities within the company that depends on the output of a previous process, while an external customer buys the product from the company. The steps of a process that do not add value from a customer point of view are waste or non-value adding (Liker, 2004). Activities that are typically considered value adding are those which transform raw materials or components into finished products. This includes assembling, forging raw materials and painting (Hines and Rich, 1997).

Lean production systems focus on elimination of waste, which is “any activity for which the customer is not willing to pay” (Dennis, 2002, p. 20). This is linked to value. Anything that does not add value is waste. Bicheno (2004, p. 14) claims that “waste prevention is at least as important as waste elimination”. This means that a company should not only remove waste, but also focus on not adding any waste to a process.

The Japanese word for waste is muda. There are two types of muda: type 1 and type 2:

Type 1 muda: activities that do not create value but are necessary in the process.

Type 2 muda: activities that do not create value and should be eliminated.

(Bicheno, 2004)

(23)

14 Type 1 muda includes walking long distances to pick up parts, unpacking deliveries or

moving a tool from one hand to the other. Activities in this category will most likely be difficult to eliminate without large changes to the system, and such changes might not be possible to achieve in the beginning. Examples of type 2 muda would be waiting time and double handling. The focus should be on those activities, seeking to eliminate them completely (Hines and Rich, 1997).

Waste is the converse of value, and it is essential to both enhance value and remove waste in order to improve. Figure 2 below shows how the steps of a process can be divided into value added activities and non value added activities. A process may often spend more time on non-value added activities, and here the focus should be on removing the unnecessary activities.

Figure 2 Illustration of value added and non-value added activities (MacMahon, 2009)

2.2.2. 7 wastes

Toyota has identified seven types of non-value adding waste, which are described in Bicheno (2004) and Liker (2004).

(24)

15 1. Overproduction.

Overproduction means producing too much or producing too early. Both create waste (Shingo, 1988). Overproduction is, according to Taiichi Ohno, also the most serious of the wastes as it is a cause of most of the other wastes. It leads to excess inventory, which consecutively causes unnecessary transport. Because of the high inventory levels and buffers, defects may not be discovered before at a late stage, and the motivation for workers to continuously improve activities might be reduced since the consequences of a machine breakdown are low when inventories are high.

2. Waiting (time on hand).

Waiting is an impediment to smooth flow. Workers may wait for work, or they might wait for a machine to finish, a tool to become available or a part to arrive. Either way it is considered waste, and the time should be spent doing something else, such as cleaning, checking or maintenance. Materials and operations may also be waiting. Materials waiting in queues and bottleneck operations waiting for work are also wastes and should be reduced.

3. Unnecessary transport or conveyance.

The movement of materials, work-in-process or finished products between processes or into or out of storage is considered a waste. It is however impossible to eliminate, so a company should aim at reducing it. An increasing number of transport and handling operations is increasing the likelihood of goods being damaged. Furthermore, the distance of

transportation affects the communication negatively. The longer distance, the harder it is to receive feedback if the quality is poor.

4. Overprocessing or incorrect processing

Overprocessing is to process a product more than necessary in order to get the desired result. Incorrect processing refers to a process which inevitably leads to the production of defects. It is a result of not having the correct methods, tools, standards, product design and training.

(25)

16 5. Excess inventory

Keeping too much inventory is a waste that both costs money and hides other problems in the process. A well-known metaphor in Lean says that inventory is the water that hides problems in the form of rocks on the bottom of the river. Only by lowering the water level, the rocks will be exposed and needed to be solved. However, the rocks must be removed before the water level is lowered, or else the ships, representing shipments, will hit the rocks and sink (Baudin, 2004). The problems, or “rocks”, may be defects and other quality problems, machine downtime, long setup times or late deliveries from suppliers. Excess inventory needs storage space which consequently increases storage costs. It also increases the risk of obsolescence, damages and delays.

6. Unnecessary movement

A non-optimal layout of a workstation leads to unnecessary movement. Workers having to walk to reach a tool or to get to another area is waste. So is looking for or stacking tools or parts. This is much about ergonomics of the workplace and also concerns health and safety.

7. Defects

Defects may be internal failure, causing scrap, rework or delay, or external failure, causing warranty, repairs, field service and maybe even a lost customer. This is a waste of handling, time and effort that could otherwise be used in value-added activities. The cost of a defect increases the longer it remains undetected; the first part with a defect may be inexpensive to correct or scrap, but if the part is connected to a finished product which is sold to a final customer, the cost will be much higher.

2.3. Lean tools

2.3.1. 5S

In the 1970s and 80`s the Americans started to visit Japanese plants to see how things worked there. What they saw were factories so clean that one could eat off the floor (Liker, 2004). Bicheno (2004) describes the five S system as a basic housekeeping system. It is also

(26)

17 designed to create a visual workplace. Dennis (2002) describes this as a workplace which is self- explaining, self-ordering and self-improving. In this kind of workplace the employees can easily see if anything is out of order, and therefore easily correct it. Here are the five S`s as they are described by Liker (2004), Bicheno (2004) and Dennis (2002):

 Sort (Seiri). The first step is to go through all the items to classify which items are needed in the workplace. Items that are not needed in the workplace should be thrown out.

 Set in order or Straighten (Seiton). The next step is to locate the items that are used in the best place. Items should be located so that they reduce the waste (muda) of motion.

 Shine (Seiso). Continue to keep the workplace clean, always look for items that are out of place. This process also helps to inspect and look for failures.

 Standardize (Seiketsu). In this step companies need to develop a system and procedures that helps monitoring and maintaining the three first steps.

 Sustain (Shitsuke). In this last step everyone should participate to maintain and continue to improve the workplace.

As shown in figure 3 the five S’s create continuous improvements in the work environment.

Liker (2004) explains that Lean systems use 5S to support a smooth flow to takt time. It can be used to make problems more visible, but it can also be part of the process of visual control of a well-planned Lean system.

(27)

18

Figure 3the 5 S’s (adapted from Liker, 2004)

2.3.2. 4M

Another tool may be the 4M checklist (sometimes referred to as 5M, in which measurement is also part of the checklist). The four M’s are:

 Man

 Machine

 Material

 Method

 (Measurement)

The 4 M’s are the inputs to a process of creating a certain level of output that is desired by the customer, as depicted in figure 4. Under each M there are ten questions that can be asked in order to identify the root causes of the problem (Imai, 1986 and Dennis, 2002).

Today all businesses have the ability to hire the same workers, buy the same material and use the same machines as their competitors. According to Keller (2010) companies can have the ability to differentiate from their competitors through their methods. The methods that can be differentiated are designing and manufacturing the products, managing orders

(28)

19 through customer service as well as selling and distributing. This is not done the same way in every business, so this may be unique for a particular company.

Figure 4A System View (adapted from Dennis, 2002)

2.3.3. 5 whys

The “5 whys” is a method for discovering the root cause of a problem or a defect, so that one can improve the cause and keep the problem or defect from recurring (Womack, Jones and Roos, 2007). Solving root causes is fundamental to the Lean philosophy. Solving root causes means that the problem is solved at the root instead of at the superficial or immediately obvious levels (Bicheno, 2004).

The reason why the technique is called the “5 whys” is because the inventor, Toyota, experienced that “why” must be asked successively five times before the root cause is established (Bicheno, 2004).

According to Bicheno (2004) many people believe that the reason why the Japanese motor industry has great quality, reliability and productivity is because of the unrelenting seek for root causes.

2.3.4. Cause- and- Effect Diagram

A useful tool to identify and systematize root causes is the cause-and-effect diagram.

Because of the shape of the diagram, see figure 5, it is also known as the fish bone diagram

(29)

20 or Ishikawa diagram (Goldsby and Martichenko, 2005). Cause-and-effect diagrams can be used in brainstorming to examine factors that may be causes of the problem. By using the tool it can be easier to narrow down the root causes of the problems (Walton, 1986).

Figure 5Cause-and-effect diagram (Management Systems Inc, 2006)

Walton (1986) mentions several benefits that can be obtained from cause-and-effect diagrams:

1. When making the diagram, discussions between the different members take place, hence people can learn from each other through discussing. The creation process can therefore be seen as educational.

2. Further the group is focused on the issue, which reduces complaints and irrelevant discussions.

3. Another benefit is that there is an active search for the cause.

4. Data must often be collected.

5. The cause-and-effect diagram can also show the level of understanding within the company. When the diagram is complex, it means that the workers are sophisticated about the process.

6. The diagram can be used for any problem.

The cause-and-effect diagram does not indicate what the right cause is, but it helps to develop educated guesses on focus measurements and finding the root causes (Pande and Holpp, 2002). By using the “5 whys” combined with a cause-and-effect diagram the root

(30)

21 causes can be narrowed down easier. Also 4M can be used in the diagram to classify the causes.

2.3.5. Value stream mapping

Value stream mapping (VSM) is a mapping tool designed to enable management to:

 Visualise the process

 Point to problems

 Focus the direction of its Lean transformation

The purpose of value stream mapping is not only to visualise how the organisations acts today, but also how they should act in the future. An advantage of value stream mapping is that it shows the big picture, so that it is easier for the company to identify the critical areas.

This is where the Lean efforts should be focused (Keyte and Locher, 2004).

Another purpose of value stream mapping is to identify opportunities for Lean

improvement. The map includes all activities within a defined process as well as the inputs to and the outputs of the process. A value stream mapping process consists of four phases, as presented by Plenert (2007):

 Preparation

 Current state map

 Future state map

 Improvement plan

These are described in detail below.

Preparation

In the preparation phase one has to identify limits and ranges of the system that is to be studied further. This implies deciding which process has highest importance (i.e. largest impact on the business) and most problems. If the limits are too narrow, one may not include something that can be improved greatly and that should be included. On the other

(31)

22 side, if the limits are too wide, it will be difficult to see the problems where the process could be improved.

Current state map

In this phase the goal is to draw a current state map describing the current situation in the company. The starting point here should be to define “value” from the customer’s point of view. Keyte and Locher (2004) also include that the main processes should be identified at this step. This is useful because it helps define the level of detail the process mapping should have. They also suggest collecting customer information such as who the customers are, their demands and their expectations of lead time.

The next step is to select appropriate key performance indicators (or process metrics). It is important to choose not too many, but just a few that are suited for the process that is being studied. Keyte and Locher (2004) emphasize that time (process time as well as lead time) always should be included as a process metric. Willoch (2005) states that a good process metrics portfolio should consist of metrics describing costs, quality and time.

Perhaps the most important step when drawing a current state map is to observe the process and perform a walk-through. This will be the basis for creating a value stream map containing specific icons and information about activities in the process. The reason for using standardized icons is that anyone who knows them can look at any map and be able to read and understand the information it contains. The value stream mapping icons used in this thesis are depicted in figure 6. By measuring the time it takes to perform the different activities as well as the time spent in between the activities, one can find the lead time of the process. The results of these measurements are necessary information when dividing all the activities into value-adding and non-value adding. The non-value adding time gives opportunities for improvement.

In this phase we should also calculate takt time, which is the time it should take to produce one unit in order to cover the customer need. The formula is:

(32)

23 Available time can be measured in hours, minutes, seconds etc, while customer demand can be number of units. The current state map helps visualising the process so that it is easier to see where the company should focus its attention.

Figure 6 Value Stream Mapping Icons (adapted from an overview in Keyte and Locher (2004))

Future state map

The future state map can be considered as the goal for the results of the Lean improvement.

It shows the ideal state that the company should try to achieve. However, since there in most cases are limited resources, the ideal state may not be achievable. So, the future state must be modified to illustrate an obtainable state. According to Keyte and Locher (2004)

(33)

24 there will always be more than one possible future state VSM, so the ones who map should choose the alternative that is best suited to the company’s goals and that are possible to achieve within a specific time frame (for example three to six months). By comparing the current state VSM and the future state VSM one can see the differences and the

improvements that should be made in the process to get to the future state. This results in an improvement plan.

Improvement plan

Here an action item list of improvements should be developed. The items on the list could be classified and ranked according to criteria such as:

- How hard is it to implement the change?

- What is the impact of the change on the process under study?

- What is the cost?

- What is the time span for implementation of the change?

- How is the item related to top management’s priorities for this Lean activity?

The ranked list is used to identify Lean “events”, which are improvement actions that a company should initiate in order to achieve the desired future state (Plenert, 2007).

2.4. Six Sigma

The Six Sigma measurement standard in product variation goes back to the 1920s (Karlöf and Lövingsson, 2005). The Six Sigma theory originates from Motorola Inc. in the United States. The company faced threats from the Japanese electronics industry in the mid eighties, and therefore needed to make drastic changes to improve their quality levels (Linderman et. al. 2003).

Harry and Schroeder (2005, p. vii) define Six Sigma as “a business process that allows companies to improve their bottom line by designing and monitoring everyday business activities in ways that minimize waste and resources while increasing customer satisfaction”.

This is about improving profitability, quality and efficiency. Six Sigma emphasizes the

(34)

25 importance of having measurements, or metrics, on what is of value to the companies. It claims that what is not measured cannot be improved.

The Six Sigma term is related to the normal distribution. Here the values are centred around the mean and then the curve flattens out symmetrically on each side of the mean. Sigma, or the standard deviation, is the distance between the mean and the inflection point of the curve, and 68 % of all data is located within one sigma to the left of the mean and one sigma to the right of the mean. As the range is expanded to two sigma, three sigma and so on, a larger share of the data is covered. This is referred to as the sigma level; the larger the share of data in the distribution that are without defects, the higher the sigma level is. At a Six Sigma level, 99.9997 % of all output is without defects, and in this context a defect is any product that does not meet customer specifications (George, 2003).

A literature review made by Tjahjono et al. (2010) has identified four interpretations of Six Sigma that to some extent overlap. The first view describes Six Sigma as a set of statistical tools within quality management that facilitate process improvement. This interpretation aims to increase performance measures to Six Sigma level, which is called critical to quality (CTQ). According to Linderman et al. (2003) a key step in any Six Sigma improvement effort is to determine exactly what the customer requires and then to define defects in terms of their CTQ parameters. Having a Six Sigma level means that the process results in 3.4 or less defective parts per million (PPM). PPM is the main quality indicator within Six Sigma (Tjahjono et al., 2010).

The second view characterizes Six Sigma as an operational philosophy of management. The philosophy is flexible and can be applied to the whole supply chain, not only production. The third view defines Six Sigma as a business culture. In addition to the use of statistical

techniques and tools, there is a need for top management commitment in order to achieve success. It is also described as an organised structure and a belief system which guides a company in decision making and uses specialists to reach strategic goals. The fourth view defines Six Sigma as an analysis methodology that uses scientific methods. It is described to be a continuous improvement methodology, with its DMAIC process (explained further in

(35)

26 chapter 2.5.) similar to Deming’s PDCA (plan, do, check, act) cycle. The Six Sigma

methodology aims to improve business processes through reducing process variability and removing waste (Tjahjono et al., 2010).

According to Pande and Holpp (2002) there are three main areas that Six Sigma targets. First of all it targets to improve customer satisfaction. It also tries to reduce cycle times, and to reduce defects. By improving these areas companies have the ability to reduce business costs and also capture new markets. Further they will retain exciting customers, but also build up a reputation for their products and services.

There are two main methods within the Six Sigma theory: DMADV and DMAIC. DMADV stands for Define, Measure, Analyze, Design and Verify. The DMADV is an improvement system that focuses on new processes or products. The method can also be used when larger improvements are needed for existing processes or products. DMAIC stands for Define, Measure, Analyze, Improve and Control (Karlöef and Löevingsson, 2005). This method is used in this thesis and will be described further in the following section.

2.5. DMAIC

DMAIC is a method within the Six Sigma methodology, which stands for: Define, Measure, Analyze, Improve and Control. According to Karlöf and Lövingsson (2005) the aim of using the DMAIC method is to improve existing processes. It was General Electrics who

introduced the four phases of measure, analyze, improve and control. Later the define phase was added (Salah, Rahim and Carretero, 2010). The stages are described further by Bicheno (2004), Pande and Holpp (2002), Goldsby and Marichenko (2005):

- Define: The main stage of the define stage is to define clearly and succinctly what the problem is. There are several sub stages within the define stage. The sub stages are to define the scope of the project, and what is important to the customer.

- Measure: How are we doing? The sub stages within the measure stage are to determine what to measure and how to measure it. Further the current performance should be

(36)

27 quantified and the improvement target should be estimated. There are three categories of measures in a process:

1. Output or outcome: the output is the result of the process. Here either the

immediate result or the long term impacts are measured. A measurement for the immediate result can for example be deliveries, complaints or defects. Measuring the long term impacts can be profit or satisfaction.

2. Process: these measurements may help find out the causes of a problem. Examples can be training hours or costs per unit.

3. Inputs: inputs are things coming into the process which are changed into outputs.

Measures can be order volume, on-time delivery and order type.

Measures that are used should be quantifiable and easy to measure. The measurements should also be robust, reliable and valid.

- Analyze: Find out what is wrong? The sub stages in the analysis will identify the causes of variation and defects. Further statistical evidence that causes are real should be provided in the analysis. One of the principles of using DMAIC is that all kinds of causes should be

considered when solving the problem. It is important that the right type of tool is used when analyzing.

- Improve: Improve what is wrong. The improve stage involves determining the solution of the problem. Then the solutions should be installed, and finally statistical evidence should be provided to show that the solutions work.

- Control: Sustain the gain that is achieved. Here controls are put into place so that

improvements are sustained over time. Further statistical evidence of sustainment should be provided.

Pande and Holpp (2002) list seven points that make DMAIC different or better than other methods:

1. Measure the problem: When using DMAIC the company cannot assume that they know what the problem is, they need to prove what the problem is with facts.

(37)

28 2. Focus on the customer: even if the company is trying to cut costs in a process the

customer is important.

3. Verify root cause: root causes must be verified with facts and data.

4. Break old habits: real changes and results can take new creative solutions.

5. Manage risks: to get rid of problems tests are done to make the solutions more perfect.

6. Measure results: verify the real impact of solutions based on facts 7. Sustain change: making changes last is important

Figure 7 shows the five different stages within the DMAIC methods. According to Goldsby and Marichenko (2005) DMAIC is the backbone of Six Sigma. The method offers a map to improve projects from the conception to the completion.

Figure 7 DMAIC model (Lean Sigma institute, 2010)

2.6. Lean and Six Sigma in service and office processes

The terms Lean and Six Sigma do not exclusively apply to manufacturing. The tools and principles may be applied to other areas of an organisation, for example service processes.

(38)

29 According to George (2003), service work may be harder to improve, due to a lack of

documented standard processes that the workers are trained in. It may then be more difficult to identify areas that need to change and how they can be improved. Office work may be less visible than physical material flow in a manufacturing process. It is therefore necessary to make this more visible, and this can be solved by making process maps and charts.

Often in office work, there is a lack of data on queues and how much time it takes to perform the tasks. Service processes are also likely to have more variation than production processes since people perform the work and not machines. It is therefore important to include people in decisions about data collection, improvement ideas and plans, and to share the results with them. Doing this will reduce any resistance to change that some people may have, at the same time as employee creativity can contribute to the plans (George, 2003).

2.7. The relationship between Lean and Six Sigma

There are several similarities and differences between Lean and Six Sigma. Salah, Rahim and Carretero (2010) have identified 5 dimensions on which Six Sigma and Lean are the same and 26 dimensions on which they are different. The 5 similar dimensions are development, leadership, principles, features and staff roles. Both concepts have their roots in Total Quality Management (TQM), and they share some of the same objectives and principles as TQM. TQM is a management system with many similarities to Six Sigma and Lean. Six Sigma and Lean have been described in this chapter, but we are not going to explain the concept of TQM in depth here. For a comparison of TQM to Lean and Six Sigma, see Andersson, Eriksson and Torstensson (2006). Six Sigma and Lean both emphasize the importance of top management commitment in the leadership dimension. Finally they both employ a project management approach with the use of team leaders and the development of improvement plans.

Some of the differences between Lean and Six Sigma include definition, complexity, focus, tools and techniques, measures and results. Salah, Rahim and Carretero (2010) claim that

(39)

30 Lean is a simpler methodology than Six Sigma and that it may be easier to understand and implement. Further the focus of the two methodologies differ as Six Sigma focuses on statistical control and defects while Lean focuses on flow and speed of products and information. However they both focus on customer satisfaction and improved financial results. The tools of Six Sigma are analytical and statistical while Lean tools are mainly analytical, but at the same time some tools are common for both methodologies. Further the measures in Lean are primarily operational and often time-based while Six Sigma measures often are financial and cost-oriented. The results in the two concepts may also be different. Examples of results within Six Sigma are reduced number of defects and higher efficiency. Results within Lean may include improvement of quality, reduction of inventory, lead time and waste.

Figure 8 The relationship between the five Lean principles and DMAIC (Salah, Rahim and Carretero, 2010)

Figure 8 shows how Six Sigma and Lean are related with regards to the Lean principles and the DMAIC methodology. The first principle, which is to identify value from the customer’s point of view, is also included in the define stage. Mapping of current processes includes measuring and collection of data which is the base for the analysis. In the improve stage efforts are being made to make the process flow better and to move towards a pull system.

The perfection principle is related to the control stage where controls and procedures are

(40)

31 introduced in order to maintain and develop further improvements (Salah, Rahim and Carretero, 2010).

According to George (2003), Lean primarily focuses on process speed, while Six Sigma primarily focuses on process quality. He claims that looking at Lean and Six Sigma as competing concepts is contradictive as one cannot achieve maximum speed without also improving quality and vice versa. Further it is argued that Lean and Six Sigma are

complimentary concepts that work well together. There are gaps in the Lean methodologies which Six Sigma can fill while Lean covers areas that Six Sigma does not. Six Sigma

emphasizes reduction of variation in processes and provides additional tools for statistical process control that Lean does not have. The key is to use Lean and Six Sigma

simultaneously in order to remove waste and improve the process.

Today the complimentary relationship between Lean and Six Sigma is accepted, and a number of companies have initiated programs that integrate the two concepts. Such programs are called Lean Six Sigma. This new term is described as a methodology that aims at eliminating both waste and variability, and it uses the DMAIC method to achieve process improvement, customer satisfaction and improvement of financial results. Both

methodologies can be regarded as tool boxes with some common tools. This gives the user a large variety of tools from which she can choose the most appropriate depending on the problem to be solved (Salah, Rahim and Carretero, 2010).

An integration of Lean and Six Sigma can use DMAIC as a structure since this is a well-known and understood methodology. For each of its different stages there are several Lean tools that can be used. Not all of the stages of DMAIC need to be equally important, depending on the problem to be solved one or more stages may have higher importance than others (Salah, Rahim and Carretero, 2010). George (2003) gives an overview of the many tools that can be used within each stage. Some of the measure tools that can contribute to description and prioritization of processes are value stream mapping, process cycle efficiency and Pareto charts. Pareto chart is a diagram which consists of bars representing the frequency of a cause or an element of a problem and a line representing the cumulative percentage of

(41)

32 the same causes or problem elements. The bars are sorted in descending order. A Pareto chart is suitable to show whether there are a few causes that make up the largest part of the problem. Process cycle efficiency (PCE) is a critical metric for waste. This shows the share of value added time to the total lead time.

PCE of 10 % or less means that there are large opportunities for removing waste. Lockheed Martin, a major producer within the aerospace industry, estimated that 83 % of the

activities from placing a purchase order until receiving the goods could be considered non- value adding. George (2003) also explains that most processes have process cycle efficiency less than 10 %.

The analyse tools include 5 whys and cause-and-effect diagrams. Both of these may be used to describe and explore cause-and-effect relationships.

Lean and Six Sigma are by some viewed as separate concepts and methodologies while it is recognized that they are similar in many ways. Others prefer to use an integration of the two, using the term Lean Six Sigma. Both Lean and Six Sigma are well-known methodologies to improve business processes.

Improving business processes : A case study of AS Spilka Industri

Master’s degree thesis

LOG950 Logistics